kernel/trace/Kconfig

   1 #
   2 # Architectures that offer an FTRACE implementation should select HAVE_FTRACE:
   3 #
   4 config HAVE_FTRACE
   5         bool
   6
   7 config TRACER_MAX_TRACE
   8         bool
   9
  10 config TRACING
  11         bool
  12         select DEBUG_FS
  13         select STACKTRACE
  14
  15 config FTRACE
  16         bool "Kernel Function Tracer"
  17         depends on DEBUG_KERNEL && HAVE_FTRACE
  18         select FRAME_POINTER
  19         select TRACING
  20         select CONTEXT_SWITCH_TRACER
  21         help
  22           Enable the kernel to trace every kernel function. This is done
  23           by using a compiler feature to insert a small, 5-byte No-Operation
  24           instruction to the beginning of every kernel function, which NOP
  25           sequence is then dynamically patched into a tracer call when
  26           tracing is enabled by the administrator. If it's runtime disabled
  27           (the bootup default), then the overhead of the instructions is very
  28           small and not measurable even in micro-benchmarks.
  29
  30 config IRQSOFF_TRACER
  31         bool "Interrupts-off Latency Tracer"
  32         default n
  33         depends on TRACE_IRQFLAGS_SUPPORT
  34         depends on GENERIC_TIME
  35         select TRACE_IRQFLAGS
  36         select TRACING
  37         select TRACER_MAX_TRACE
  38         help
  39           This option measures the time spent in irqs-off critical
  40           sections, with microsecond accuracy.
  41
  42           The default measurement method is a maximum search, which is
  43           disabled by default and can be runtime (re-)started
  44           via:
  45
  46               echo 0 > /debugfs/tracing/tracing_max_latency
  47
  48           (Note that kernel size and overhead increases with this option
  49           enabled. This option and the preempt-off timing option can be
  50           used together or separately.)
  51
  52 config PREEMPT_TRACER
  53         bool "Preemption-off Latency Tracer"
  54         default n
  55         depends on GENERIC_TIME
  56         depends on PREEMPT
  57         select TRACING
  58         select TRACER_MAX_TRACE
  59         help
  60           This option measures the time spent in preemption off critical
  61           sections, with microsecond accuracy.
  62
  63           The default measurement method is a maximum search, which is
  64           disabled by default and can be runtime (re-)started
  65           via:
  66
  67               echo 0 > /debugfs/tracing/tracing_max_latency
  68
  69           (Note that kernel size and overhead increases with this option
  70           enabled. This option and the irqs-off timing option can be
  71           used together or separately.)
  72
  73 config SCHED_TRACER
  74         bool "Scheduling Latency Tracer"
  75         depends on DEBUG_KERNEL
  76         select TRACING
  77         select CONTEXT_SWITCH_TRACER
  78         select TRACER_MAX_TRACE
  79         help
  80           This tracer tracks the latency of the highest priority task
  81           to be scheduled in, starting from the point it has woken up.
  82
  83 config CONTEXT_SWITCH_TRACER
  84         bool "Trace process context switches"
  85         depends on DEBUG_KERNEL
  86         select TRACING
  87         select MARKERS
  88         help
  89           This tracer gets called from the context switch and records
  90           all switching of tasks.
  91
  92 config DYNAMIC_FTRACE
  93         bool "enable/disable ftrace tracepoints dynamically"
  94         depends on FTRACE
  95         default y
  96         help
  97          This option will modify all the calls to ftrace dynamically
  98          (will patch them out of the binary image and replaces them
  99          with a No-Op instruction) as they are called. A table is
 100          created to dynamically enable them again.
 101
 102          This way a CONFIG_FTRACE kernel is slightly larger, but otherwise
 103          has native performance as long as no tracing is active.
 104
 105          The changes to the code are done by a kernel thread that
 106          wakes up once a second and checks to see if any ftrace calls
 107          were made. If so, it runs stop_machine (stops all CPUS)
 108          and modifies the code to jump over the call to ftrace.
 109
 110 config FTRACE_SELFTEST
 111         bool
 112
 113 config FTRACE_STARTUP_TEST
 114         bool "Perform a startup test on ftrace"
 115         depends on TRACING
 116         select FTRACE_SELFTEST
 117         help
 118           This option performs a series of startup tests on ftrace. On bootup
 119           a series of tests are made to verify that the tracer is
 120           functioning properly. It will do tests on all the configured
 121           tracers of ftrace.